| 1. |
1P1-M05 6DOF-SLAM using Omnidirectional Stereo Vision with Scan-Matching(Vision System for Mobile Robot)
This paper proposes a method to perform 6-DOF SLAM by using scan matching based on the three-dimensional distance information acquired by omnidirectional stereo vision. Acquired scan is evaluated by the Iterative Closest Point (ICP) algorithm, then the result is integrated by the extended Kalman filter (EKF). Evaluation of the proposed method is done by the experiments on a simulation environment. The evaluation results show the effectiveness of the proposed method.. |
| 2. |
A Two-Dimensional Code with Auxiliary Lines and its Decode Method for Local Distortion Correction
General 2D code decoders read 2D codes without any distortion or with uniform distortion, and it is hard for the decoders to read 2D codes involving partial, non-linear distortion. This paper proposes a 2D code involving colored auxiliary lines and its decoder. The proposed decoder can read the proposed 2D codes by using the auxiliary lines based on de Bruijn sequence, even if they are distorted locally and non-linearly. The proposed decoder also detects self-occlusion by the de Bruijn sequence, and supplements the self-occluded area due to its distortion by error correction of a QR code.. |
| 3. |
DS-1-2 Comments Recommendations for Images Based on Similarity of Images and Semantic Features. |
| 4. |
A Comparative Study on 3-Dimensional Registration by Evolutionary Computation and Its Application to Entire Shape Reconstruction. |
| 5. |
2P1-J08 Plane Detection for Recognizing the Structure of Environment using Omnidirectional Stereo Vision(Vision System for Mobile Robot)
This paper proposes a plane detection method for recognizing the structure of environment using the three-dimensional distance information that computed by omnidirectional stereo vision. Omnidirectional stereo vision can acquire a wide field of view, therefore we can detect a many of planes that construct the surroundings. Evaluation of the proposed method is don e by the experiments on a simulation environment. The evaluation results show the effectiveness of the proposed method.. |
| 6. |
An Entire 3-Dimensional Reconstruction Method without Manual Adjustment by Evolutionary Computation. |
| 7. |
Biomedical Measurement by Applying Active Stereo Image Measurement. |
| 8. |
5-1 Reflectance Parameter Estimation for 3D Object's Surface by Evolutionary Computation
In the estimation of BRDF parameter, a simple local search method may not be able to find an optimal solution. This paper proposes a new optimization method which is a combination of case-based reasoning and selfadaptive differential evolution method. This method can find efficiently high quality reflection parameters.. |
| 9. |
Coded Aperture Optimization for 3D Measurement from Projector's Defocus
This paper proposes a method to design a coded aperture, which is installed in a projector for active depth measurement. Techniques to estimate depth from the degree of image blur are known as Depth from Defocus (DfD), and some coded apertures for DfD have been proposed. However, such coded apertures are designed for cameras and not guaranteed to work properly for the projector. Since the aperture pattern having spatial frequency characteristics suitable for DfD is not obvious, it is necessary to determine by the search. Therefore, in this papaer, we propose a method to create a coded aperture using Genetic Algorithm (GA). In particular, we define an evaluation function based on depth estimation simulation. Experiments have shown that obtained pattern by GA is better than other patterns.. |
| 10. |
1A2-I06 ICP-based Ego-motion Estimation Considering Range Ambiguity Using Omnidirectional Stereo(Localization and Mapping (1))
Range-based ego-motion estimation method, which is so-called "scan-matching", has an important role in robotics. However, iterative closest points (ICP), which is basic scan-matching method, assumes that the error of obtained range data is constant. This degrades estimation results when the robot uses sensors whose error is not constant, i.e.stereo vision. In this paper, we propose a ICP-based ego-motion estimation method considering range ambiguity of the stereo vision. The key idea is comparing obtained 3D points in not Euclidean space but disparity space. We implement the method to the robot mounting omnidirectional stereo vision. Simulation experiments are conducted to show the effectiveness of our method.. |
| 11. |
Facial motion recognition Based on Dense and Detailed Motion data
3D reconstruction methods which can measure shapes with high density and high frame-rate have been proposed due to the progress of 3D reconstruction research area. In this paper, we propose a method for recognizing micro-expressions which are tiny motions appearing on a face by using the high density and high frame-rate 3D reconstruction method. Some studies report that the micro-expressions are caused by the change of mental state. If we can recognize the micro-expressions, this information could be useful for machines to understand the mental state of a human. To detect a part of the face where the micro-expressions are appeared, we propose an experimental estimation of the part. We also report a recognition rate of the change of the mental state using the experimental system.. |
| 12. |
Super Resolution of Fisheye Camera Images with Widely Varying Resolution
Fisheye cameras are widely adopted in the back view camera of vehicles and the surveillance camera. Compared to the normal digital cameras, their resolution and image quality are low. In this paper, we propose a method to restore the high resolution images from the fisheye camera by the reconstructured type of super resolution. First, we undistort the fisheye images and regularize these images. Then we super-resolve these regularized images. Blur kernel for super resolution includes the lens blur and the expansion and reduction by the undistortion. The experimental results confirms to the effectiveness of our method which applies to the images captured by the digital camera with fish-eye lens.. |
| 13. |
Improvement of a Decode Method for Two-Dimensional Codes with Auxiliary Lines against Local Distortion
General Two-Dimensional (2D) code decoders read 2D codes without any distortion or with uniform distortion, and it is hard for the decoders to read 2D codes involving partial, non-linear distortion. The authors have proposed a 2D code involving colored auxiliary lines for the correction of distortion and recognition of occlusion regions. However, the decoder sometimes fails to read modules around an occlusion boundary. Therefore, this paper attempts to improve a decode method for the 2D code involving the colored auxiliary lines. The improved decoder revises line recognition result by estimating reliability of line recognition and iterated dynamic programming matching. Experimental results show that the proposed method decodes the 2D code involving partial, nonlinear distortion with higher accuracy than the previous decoder.. |
| 14. |
Improvement of a Decode Method for Two-Dimensional Codes with Auxiliary Lines against Local Distortion
General Two-Dimensional (2D) code decoders read 2D codes without any distortion or with uniform distortion, and it is hard for the decoders to read 2D codes involving partial, non-linear distortion. The authors have proposed a 2D code involving colored auxiliary lines for the correction of distortion and recognition of occlusion regions. However, the decoder sometimes fails to read modules around an occlusion boundary. Therefore, this paper attempts to improve a decode method for the 2D code involving the colored auxiliary lines. The improved decoder revises line recognition result by estimating reliability of line recognition and iterated dynamic programming matching. Experimental results show that the proposed method decodes the 2D code involving partial, nonlinear distortion with higher accuracy than the previous decoder.. |
| 15. |
D-12-29 Representation of surface motions of human body based on bone structure and height maps. |
| 16. |
2A2-S06 Scan Matching-based 6DOF Ego-motion Estimation with Uncertainty(Localization and Mapping)
Range-based ego-motion estimation method, which is so-called "scan-matching", has an important role in robotics. However, conventional scan-matching method, assumes that the error of obtained range data is constant. This degrades estimation results when the robot uses sensors whose error is not constant, e.g.stereo vision. Furthermore, we must also be taken into account for the error in the ego-motion were estimated. In this paper, we propose a scan matching-based 6DOF ego-motion estimation method that considering range ambiguity of the stereo vision and error in the ego-motion were estimated. Simulation experiments are conducted to show the effectiveness of our method.. |
| 17. |
MATSUSHITA Yuki, KAWASAKI Hiroshi, ONO Shintaro, IKEUCHI Katsushi, SIMULTANEOUS DEBLUR AND SUPER-RESOLUTION TECHNIQUE FOR VIDEO SEQUENCE CAPTURED BY HAND-HELD VIDEO CAMERA, 鹿児島大学工学部研究報告 = The Research Reports of the Faculty of Engineering, Kagoshima University, Vol.57, p.33, 2015.11. |
| 18. |
A Fundamental Study on Application of Coded Aperture to Digital Watermark Extraction. |
| 19. |
An experimental study on coded aperture design for digital watermark extraction. |
| 20. |
Rendering Intricately-Shaped Objects Using View-Dependent Microfacets
仮想空間内でコンピュータグラフイックスを用いて実物体を表示する際には,対象物の幾何情報とテクステャ情報を用いて描画する方法が一般的である.しかし物体が複雑な細部構造をもつ場合,幾何モデルを正確に計測することができないため,現実感を失うことなくこれらの物体を描画することが難しい.そこで描画視点に依存して視線と面が常に直交するように微小面を層状に発生させ,更に各面に視点依存テクスチャマッピングを行うことで,より現実感の高い描画を行う手法を提案する.これらの処理はプログラム可能グラフイックスハードウェアを使って通常のPCで効率的に実装することが可能である.本手法は視点依存の詳細度制御が可能であり,複雑な物体や風景を対話的な速度で描画できることを示す.. |
| 21. |
Thanda Oo, Hiroshi Kawasaki, Yutaka Ohsawa, Katsushi Ikeuchi, The separation of reflected and transparent layers from real-world image sequence, MACHINE VISION AND APPLICATIONS, 10.1007/s00138-006-0043-1, Vol.18, No.1, pp.17-24, 2007.02, This paper presents an optimal method for the separation of reflected and transparent layers from real-world scene images. Whereas past research has been applied to indoor environments and static cameras, our technique can be used for outdoor scenes and motion cameras. The method is based on spatio-temporal analysis, especially using epipolar plane images (EPI). The edge and color information of EPI has been used to segment the areas on EPIs efficiently and separate the reflected and transparent layers. This method can be used for refining building textures by removing reflections from captured images for the purpose of city modeling.. |
| 22. |
Thanda Oo, Hiroshi Kawasaki, Yutaka Ohsawa, Katsushi Ikeuchi, The separation of reflected and transparent layers from real-world image sequence, MACHINE VISION AND APPLICATIONS, 10.1007/s00138-006-0043-1, Vol.18, No.1, pp.17-24, 2007.02, This paper presents an optimal method for the separation of reflected and transparent layers from real-world scene images. Whereas past research has been applied to indoor environments and static cameras, our technique can be used for outdoor scenes and motion cameras. The method is based on spatio-temporal analysis, especially using epipolar plane images (EPI). The edge and color information of EPI has been used to segment the areas on EPIs efficiently and separate the reflected and transparent layers. This method can be used for refining building textures by removing reflections from captured images for the purpose of city modeling.. |
| 23. |
Realtime Human body scanning system using grid pattern based One-shot scanner
Dense 3D reconstruction of objects could contribute to various applications such as body structure alalysis. To achieve this, many method have proposed and especially techniques using structured light have been proposed. In addition, the techniques using a single pattern can capture dynamic scene and it has a wide range for applications. The method called oneshot-scanning. In this papers, we propose 3D reconstruction based infrared ray. Its technique using infrared rays with a particular wavelength has advantages such as be lightly affected by environment light and brightness during face measurement, and can obtain texture coincidentally. In this paper,we construct a human body scanning system using infrared oneshot scanner and evaluate this.. |
| 24. |
Hiroshi Kawasaki, Ryo Furukawa, Shape Reconstruction and Camera Self-Calibration Using Cast Shadows and Scene Geometries, INTERNATIONAL JOURNAL OF COMPUTER VISION, 10.1007/s11263-008-0188-7, Vol.83, No.2, pp.135-148, 2009.06, Recently, various techniques of shape reconstruction using cast shadows have been proposed. These techniques have the advantage that they can be applied to various scenes, including outdoor scenes, without using special devices. Previously proposed techniques usually require calibration of camera parameters and light source positions, and such calibration processes limit the range of application of these techniques. In this paper, we propose a method to reconstruct 3D scenes even when the camera parameters or light source positions are unknown. The technique first recovers the shape with 4-DOF indeterminacy using coplanarities obtained by cast shadows of straight edges or visible planes in a scene, and then upgrades the shape using metric constraints obtained from the geometrical constraints in the scene. In order to circumvent the need for calibrations and special devices, we propose both linear and nonlinear methods in this paper. Experiments using simulated and real images verified the effectiveness of this technique.. |
| 25. |
Hiroshi Kawasaki, Ryo Furukawa, Shape Reconstruction and Camera Self-Calibration Using Cast Shadows and Scene Geometries, INTERNATIONAL JOURNAL OF COMPUTER VISION, 10.1007/s11263-008-0188-7, Vol.83, No.2, pp.135-148, 2009.06, Recently, various techniques of shape reconstruction using cast shadows have been proposed. These techniques have the advantage that they can be applied to various scenes, including outdoor scenes, without using special devices. Previously proposed techniques usually require calibration of camera parameters and light source positions, and such calibration processes limit the range of application of these techniques. In this paper, we propose a method to reconstruct 3D scenes even when the camera parameters or light source positions are unknown. The technique first recovers the shape with 4-DOF indeterminacy using coplanarities obtained by cast shadows of straight edges or visible planes in a scene, and then upgrades the shape using metric constraints obtained from the geometrical constraints in the scene. In order to circumvent the need for calibrations and special devices, we propose both linear and nonlinear methods in this paper. Experiments using simulated and real images verified the effectiveness of this technique.. |
| 26. |
Efficient Projector Calibration Method using Plane with Checkerboard Pattern
In recent years, development on 3D measurement system using camera and projector has been intensively conducted and stable and accurate projector calibration technique is strongly required. Although camera calibration algorithm is now widely available, e.g. OpenCV, projector calibration is not yet. Since current projector calibration method requires special calibration box and/or multiple projections of special patterns to retrieve one-to-one correspondences, it is a time consuming and labarious task. In this paper, we propose an efficient projector calibration method using plane with checkerboard pattern so that calibration process can be simple, accurate and stable. Successful experimental results are also shown in the paper.. |
| 27. |
Efficient Projector Calibration Method using Plane with Checkerboard Pattern
In recent years, development on 3D measurement system using camera and projector has been intensively conducted and stable and accurate projector calibration technique is strongly required. Although camera calibration algorithm is now widely available, e.g. OpenCV, projector calibration is not yet. Since current projector calibration method requires special calibration box and/or multiple projections of special patterns to retrieve one-to-one correspondences, it is a time consuming and labarious task. In this paper, we propose an efficient projector calibration method using plane with checkerboard pattern so that calibration process can be simple, accurate and stable. Successful experimental results are also shown in the paper.. |
| 28. |
3D reconstruction from coplanarity constraints and bas-relief transformations
カメラで観測された 3 次元点の間に，多数の共面性の拘束が成り立つとき，どれだけ多数の共面性条件を利用しても，シーン全体には自由度が残る．この自由度は generalized perspective bas-relief (GPBR) 変換によるものであり，未校正光源による shape-from-shadow 手法におけるシーンと同じ形式の自由度でもある． GPBR 変換は，直感的には理解しにくい変換であるが，シーンの双対空間を考えると，元の空間に働く GPBR 変換は，双対空間では並進及びスケーリングとなり，より理解しやすい．本稿では，このような観点から，幾つかの問題設定についての GPBR 変換の自由度が，双対空間においてどのような変換になるかについて説明し，これを解の自由度の考察や解消に利用する方法を述べる．When multiple coplanarity constraints exists in a scene observed by a camera, there remains 4 degrees of freedom in the scene that preserves all the coplanarity constraints. The freedom of the solutions is based on generalized perspective bas-relief(GPBR) transformations, which are also known as the ambiguity of shape-from-shadow method using uncalibrated light sources. Although GPBR transformations are not intuitive transformations, their dual transformations that work in the dual space are translations and scaling, which are more intuitive. In this paper, we explain the ambiguity of GPBR transformations of some problems from the dual space. For some problems, the discussion is used to explain geometrical meanings of ambiguity of the solutions, and to solve the ambiguity.. |
| 29. |
Ryo Furukawa, Hiroshi Kawasaki, Grid-based oneshot scan using dot-line pattern, 3D Image Acquisition and Display: Technology, Perception and Applications: OSA imaging congress, 2018.06. |
